WSP12650 - Laserfiche WebLink

Home Browse Search

UUlao, SAMPLING VARIABILITY FOR RARE RIVERINE FISHES 1647 The mean CV, however, did not differ among seasons (F2,6, P=0'53) or between diel periods (F1,6, P= 0,16). There was higher variability in CPUE of humpback chub ~200 mm Lr using electrofishing, regardless of season or diel period. There were no two or three-way interactions among effects (i.e. diel period, gear type and season; P ~ 0.40) in the tlannelmouth sucker ANCOV A of the CV. In addition, mean CV did not differ among seasons (F2,6, P=0.53), between diel periods (F1,6, P = 0.40), but tended to differ between gear types (F1,6, P=0'12) (Table I). There tended to be higher variability in CPUE of flannelmouth suckers ~200 mm Lr by electrofishing compared to trammel netting, regardless of season or diel period. An analysis of gear types could not be conducted on bluehead sucker CPUE because of electrofishing produced only one bluehead sucker ~200 mm Lr. The CPUE comparisons, however, were made between diel periods and among seasons for trammel nets. There was no difference in mean CV of bluehead suckers among seasons (F2,2, P=0'95) or diel periods (F1,2, P=0.90), or the interaction between season and diel period (F1,2> P=0'50). Very few trammel nets, however, collected bluehead suckers (278 of 303 trammel net sets had no fish). CPUE RELATIONSHIPS IN 'PAIRED' SAMPLES Trammel net and electrofishing CPUE for samples collected at the same location, date and hour were not positively related for humpback chub (r = -0' 31, n = 9, P = 0'43) and flannelmouth suckers (r = -0'27, n = 10, P = 0'46). In many instances, one gear collected fishes whereas the other gear collected no fishes in these 'paired' samples. Bluehead sucker CPUE compari- sons were not made because only one sample collected at least one bluehead sucker in the trammel net and electro fishing. Clearly, CPUE of native fishes from trammel nets does not necessarily reflect the CPUE of electro fishing. ESTIMA TES OF SAMPLE SIZES REQUIRED Trammel netting required lower sample sizes to detect changes in CPUE for both humpback chub and flannelmouth suckers. To detect a 10% change in CPUE with a power of 0'9, however, trammel netting still required up to 3844 net sets for humpback chub and up to 5279 for flannel mouth suckers (Fig. I). Sample sizes required to detect a 25% change in CPUE for humpback chub ranged from 170 (trammel netting at power = 0.6) to 2944 samples (electrofishing at power = 0'9). To detect a 25% change in flannelmouth sucker CPUE, a minimum of 304 samples were need for trammel netting at a power = 0.6 (Fig. I). The samples size needed for trammel netting bluehead suckers ranged from 489 (25% change at 0,6 power) to 8521 (10% change at 0.9 power). In general, samples sizes to detect changes in CPUE in this Colorado River reach were lower for trammel nets, but still constituted a very high and logistically impractical amount of effort that probably would sample the entire reach completely. Il) 2004 The Fisheries Society of the British Isles, Journal of Fish Biology 2004, 65, 1643-1652 (No claim to original US government works)